The long-term goal of this program of research is to describe and to understand changes in hearing that occur during human infancy. Recent work has focused on infants' ability to process one sound in the presence of competing sounds. The results show that the sound properties that influence this ability differ between adults and 7-9-month-old infants, suggesting that the mechanisms underlying listening in complex environments are also different. A new synthesis of work describing the development of auditory neural pathways leads to new hypotheses concerning the nature of such age-related changes in hearing. Specifically, it appears that the source of input to auditory cortex shifts around 6 months of age. If that is the case, then infants younger than 6 months are predicted to respond in a qualitatively different way from infants older than 6 months to situations in which selective processing of one sound in the presence of another is required. The goal of the proposed work is to test that general hypothesis. The proposed experiments address three specific aims. The first aim is to determine whether 3- and 7-month-old infants are able to extract some of the important sound attributes- pitch, spectral shape, and modulation-that are used by adults to separate competing sounds into different auditory streams. Infants' ability to discriminate vowels and musical notes that differ along these dimensions will be tested. It is hypothesized that infants will be able to discriminate sounds along all three dimensions. The second specific aim is to determine whether 3- and 7-month-old infants are able to separate two concurrent streams of vowels or musical notes. The concurrent streams will differ along the dimensions of pitch, spectral shape and modulation, according to the results of the experiments addressing the first specific aim. Infants' ability to discriminate a change in a vowel or musical note in one stream in the presence of an unchanging stream will be tested. It is hypothesized that both 3-month-olds and 7-month-olds will be able to perform this task when the streams differ in spectral shape and perhaps modulation, but not when they differ in pitch. The third specific aim is to determine whether 3- and 7-month-old infants are able to selectively process a change in a vowel or musical note in only one of two streams when the same change can occur in the competing stream. Infants' ability to discriminate a change in a vowel or musical note in a specific stream will be tested, but under conditions in which the same change sometimes occurs in the competing stream. It is hypothesized that 7-month-olds, but not 3-month-olds will be able to perform this task. The results of these experiments will have implications for current models of auditory neural development and for models of perceptual organization in mature listeners.